This project outlines the development and characterization of novel mouse models containing human variants of the brain gene, NR2E1. NR2E1 is an orphan nuclear receptor highly conserved between humans and mouse. In mouse Nr2e1 has been shown to have important roles in forebrain development and neurogenesis. Work in our lab has shown that mice lacking Nr2e1 have abnormal brain development and exhibit hyperactive and extreme aggressive behaviors. We have also shown that these phenotypes can be rescued by the human NR2E1 under its endogenous promoter, thereby demonstrating the functional equivalence of the genes in mouse. This work led us to examine NR2E1 variants in humans - and hypothesis supported by human linkage studies implicating the NR2E1 region on 6q21. Excitingly, we have discovered an association between an NR2E1 polymorphism and individuals who have bipolar disorder and identified 8 novel variants in patients that were not found in controls. Having discovered a candidate gene for bipolar disorder, the next critical step is to test these variants in functional assays such as the mouse. Importantly, our experimental plan will use a knock-in approach to generate mice with a single copy of human NR2E1 variants knocked into a specific locus at hypoxanthine- guanine phosphoribosyltransferase (Hprt). This strategy offers advantages in that Hprt knockout mice have no known phenotype and the locus allows unrestricted expression reflective of the promoter used. Most importantly, this strategy most closely models the human condition since mice will have a single copy of the variant rather than multiple copies generated by traditional random insertion approaches. A final advantage is that several variants can be directly compared as there are no site-of-insertion effects to confound phenotypic characterization. On one level, this project meets the goals of the RFA - to generate a mouse model to test human brain disease candidate genes. However, it also establishes proof of principle for an adaptable high throughput system to test other candidates. Aim 1 -- To establish the approach and demonstrate its effectiveness we will generate mice carrying two extreme human NR2E1 alleles;wild-type (Wt) and a regulatory deletion (Del). Aim 2 -- To challenge the approach with increased variant difficulty (single base-pair regulatory changes) and to test its effectiveness in generating human candidate variants we will generate mice carrying three NR2E1 variants (Var1, 2, and 3). Aim 3 -- To explore the functional relevance of these human variants to brain disease we will undertake neurophenotyping of the 5 strains generated in Aims 1 and 2;i.e. brain histology, behavior testing, and testing of the most favorable 2 strains for differential effects of the psychotherapeutic drug lithium. Bipolar disorder (manic-depressive disorder) is an incurable, long-term brain disorder affecting 5.7 million Americans. The genetics and disease mechanisms of bipolar disorder are poorly understood and effective, well-tolerated therapeutic options are greatly lacking. Studies of the effects of human mutations of the brain gene, NR2E1, in mice will provide new insight into our understanding of bipolar disorder and ultimately enable the development of new diagnostic and therapeutic advances.